Adding a "quality" or "confidence" factor to the dataset

[marshallmcdonnell]

Would like to have a field to annotate either the quality of the dataset (specifically, I am looking at the dataseries entry) or a "confidence" in this dataset.

The idea is to use this later for filtering based on quality in SPARQL queries to either include more data with "less quality" or less data with "better quality".

Currently, my planned steps to implement will be:

• [ ] Investigate what ontological type would best fit this field to semantically describe it (good resource for search: https://www.ebi.ac.uk/ols/index). Preferably use one already available if it is a good fit in the context files.
• [ ] If there is a requirement for a new entry into the context for this type, submit a PR for setting up a "development" context file that is included so we can associate the new field with this ontological type. If already included in current context, skip this step
• [ ] Mock up SciData JSON-LD files with this new field for the "quality" of the dataset (or the dataseries, if better to put it "deeper") for a few different quality values (mix of low and high for testing filtering; i.e. 25%, 50%, 99%, etc.)
• [ ] Upload to the graph database (currently, we are using Apache Jena with TDB2)
• [ ] Test out SPARQL queries to see if we can indeed filter based on this value
• [ ] Submit PR for this new entry into SciData framework
  • [ ] Add it to the necessary files:
    • [ ] https://github.com/stuchalk/scidata/blob/master/contexts/scidata_dataset.jsonld
    • [ ] https://github.com/stuchalk/scidata/blob/master/examples/sections/dataset.jsonld
    • [ ] https://github.com/stuchalk/scidata/blob/master/examples/sections/dataseries.jsonld
    • [ ] https://github.com/stuchalk/scidata/blob/master/examples/sections/framework.jsonld
    • [ ] https://github.com/stuchalk/scidata/blob/master/examples/sections/framework_plus.jsonld